paper-with-me

Papers

MoDA: Modulation Adapter for Fine-Grained Visual Grounding in Instructional MLLMs

2025-06-02 · Wayner Barrios, Andrés Villa, Juan León Alcázar, SouYoung Jin, Bernard Ghanem

Recently, Multimodal Large Language Models (MLLMs) have demonstrated impressive performance on instruction-following tasks by integrating pretrained visual encoders with large language models (LLMs). However, existing approaches often struggle to ground fine-grained visual concepts in complex scenes. In this paper, we propose MoDA (Modulation Adapter), a lightweight yet effective module designed to refine pre-aligned visual features through instruction-guided modulation. Our approach follows the standard LLaVA training protocol, consisting of a two-stage process: (1) aligning image features to the LLMs input space via a frozen vision encoder and adapter layers, and (2) refining those features using the MoDA adapter during the instructional tuning stage. MoDA employs a Transformer-based cross-attention mechanism to generate a modulation mask over the aligned visual tokens, thereby emphasizing semantically relevant embedding dimensions based on the language instruction. The modulated features are then passed to the LLM for autoregressive language generation. Our experimental evaluation shows that MoDA improves visual grounding and generates more contextually appropriate responses, demonstrating its effectiveness as a general-purpose enhancement for image-based MLLMs.

📄 PDF Abstract BibTeX arXiv:2506.01850

Code (0)

등록된 구현이 없습니다.

Tasks

Instruction FollowingText GenerationVisual Grounding

Methods 이 논문이 사용한 방법론

Adapter 설명 없음

Similar Papers 제목 키워드 기반

MQAdapter: Multi-Modal Quantum Adapter for Coarse-to-Fine VLM Fine-tuning

2026-07-14 · Yumiao Zhao, Bo Jiang, Min Lu, Xiao Wang 외 arxiv

Large-scale Vision-Language Models have demonstrated impressive transfer learning capabilities across a wide range of tasks. For few-shot classification, we observe that VLMs exhibit a notable ability to filter candidate…

Few-Shot LearningTransfer Learning

HiVG: Hierarchical Multimodal Fine-grained Modulation for Visual Grounding

2024-04-20 · Linhui Xiao, Xiaoshan Yang, Fang Peng, YaoWei Wang 외

Visual grounding, which aims to ground a visual region via natural language, is a task that heavily relies on cross-modal alignment. Existing works utilized uni-modal pre-trained models to transfer visual or linguistic k…

cross-modal alignmentVisual Grounding

Fine-Grained Scene Image Classification with Modality-Agnostic Adapter

2024-07-03 · Yiqun Wang, Zhao Zhou, Xiangcheng Du, Xingjiao Wu 외

When dealing with the task of fine-grained scene image classification, most previous works lay much emphasis on global visual features when doing multi-modal feature fusion. In other words, models are deliberately design…

image-classificationImage Classification

TaxaAdapter: Vision Taxonomy Models are Key to Fine-grained Image Generation over the Tree of Life

2026-03-27 · Mridul Khurana, Amin Karimi Monsefi, Justin Lee, Medha Sawhney 외 arxiv

Accurately generating images across the Tree of Life is difficult: there are over 10M distinct species on Earth, many of which differ only by subtle visual traits. Despite the remarkable progress in text-to-image synthes…

Image Generation

Fine-Grained VLM Fine-tuning via Latent Hierarchical Adapter Learning

2025-08-15 · Yumiao Zhao, Bo Jiang, Yuhe Ding, Xiao Wang 외 arxiv

Adapter-based approaches have garnered attention for fine-tuning pre-trained Vision-Language Models (VLMs) on few-shot classification tasks. These methods strive to develop a lightweight module that better aligns visual …

Few-Shot Learning